201805032 - Ardıl Silan Aydın 201805016 - Beril Kartal 201805060 - Elif Yılmaz 201805017 - Neslihan Özdil 201805045 - Orhan Gazi Barak 201805040 - Yusuf Çelikkıran

Checking and improving column names

import pandas as pd

# Verisetini yükle
data = pd.read_csv("isedataset.csv")

# Sütun isimlerini yazdır
print(data.columns)

Index(['Unnamed: 0', 'Open', 'High', 'Low', 'Close', 'Volume', 'Symbol',
       'Predict', 'Unnamed: 8'],
      dtype='object')

import pandas as pd

# Verisetini yükle
data = pd.read_csv("isedataset.csv")

# Sütun isimlerini belirle
data.columns = ['Date', 'Open', 'High', 'Low', 'Close', 'Volume', 'Symbol', 'Predict','Predict2']
# Sütun sil
data.drop("Predict2", axis=1, inplace=True)

# Sütun isimlerini yazdır
print(data.columns)
data.to_csv("isedataset_uptade.csv", index=False)

Index(['Date', 'Open', 'High', 'Low', 'Close', 'Volume', 'Symbol', 'Predict'], dtype='object')

Converting data to numeric

import pandas as pd
from sklearn.preprocessing import LabelEncoder

# Veri setini yükle
data = pd.read_csv("isedataset_uptade.csv")

# Tarih sütununu datetime nesnesine dönüştür
data['Date'] = pd.to_datetime(data['Date'])

# Tarih özelliklerini ayır
data['Year'] = data['Date'].dt.year
data['Month'] = data['Date'].dt.month
data['Day'] = data['Date'].dt.day

data.drop("Date", axis=1, inplace=True)

# LabelEncoder oluştur
label_encoder = LabelEncoder()

# Symbol sütununu label encoding uygula
data['Symbol_Encoded'] = label_encoder.fit_transform(data['Symbol'])
data.drop("Symbol", axis=1, inplace=True)

# "Predict" sütununu veri setinden çıkar
predict_column = data.pop("Predict")

# "Predict" sütununu en sona ekle
data["Predict"] = predict_column

# Sonucu yazdır
print(data.head())

# Veriyi yeni CSV dosyasına kaydet
data.to_csv("Numeric_isedataset.csv", index=False)

        Open       High        Low      Close    Volume  Year  Month  Day  \
0  27.500000  27.500000  27.500000  27.500000    262214  2023      6   15   
1  30.240000  30.240000  30.240000  30.240000   1169499  2023      6   16   
2  31.000000  31.100000  29.940001  29.940001   8064437  2023      6   20   
3  26.959999  26.959999  26.959999  26.959999   2147415  2023      6   21   
4  25.620001  27.620001  25.500000  25.940001  71898180  2023      6   22   

   Symbol_Encoded    Predict  
0               0  30.240000  
1               0  29.940001  
2               0  26.959999  
3               0  25.940001  
4               0  25.900000  

Creating new features

import pandas as pd

# Veri setini yükle
data = pd.read_csv("Numeric_isedataset.csv")

# Hareketli Ortalama (Moving Average) hesaplama
data['5_Day_Average'] = data.groupby('Symbol_Encoded')['Close'].transform(lambda x: x.rolling(window=5, min_periods=1).mean())
data['10_Day_Average'] = data.groupby('Symbol_Encoded')['Close'].transform(lambda x: x.rolling(window=10, min_periods=1).mean())

# Günlük Fiyat Değişimi (Daily Price Change) hesaplama
data['Daily_Price_Change'] = data.groupby('Symbol_Encoded')['Close'].transform(lambda x: x.pct_change())

# Hacim İstatistikleri hesaplama
data['Average_Volume'] = data.groupby('Symbol_Encoded')['Volume'].transform(lambda x: x.rolling(window=5, min_periods=1).mean())
data['Maximum_Volume'] = data.groupby('Symbol_Encoded')['Volume'].transform(lambda x: x.rolling(window=5, min_periods=1).max())
data['Minimum_Volume'] = data.groupby('Symbol_Encoded')['Volume'].transform(lambda x: x.rolling(window=5, min_periods=1).min())

# Volatilite hesaplama
data['Volatility'] = data.groupby('Symbol_Encoded')['Close'].transform(lambda x: x.rolling(window=5, min_periods=1).std())

# Önceki gün kapanış fiyatı
data['Previous_Day_Close'] = data.groupby('Symbol_Encoded')['Close'].shift(1)

# Relatif Güç Endeksi (Relative Strength Index - RSI)
def rsi(values):
    delta = values.diff()
    gain = (delta.where(delta > 0, 0)).rolling(window=14, min_periods=1).mean()
    loss = (-delta.where(delta < 0, 0)).rolling(window=14, min_periods=1).mean()
    rs = gain / loss
    return 100 - (100 / (1 + rs))

data['RSI'] = data.groupby('Symbol_Encoded')['Close'].transform(rsi)

# NaN değerleri temizleme
data.dropna(inplace=True)

# Sonucu yazdır
print(data.head())

# Veriyi yeni CSV dosyasına kaydet
data.to_csv("Numeric_isedataset_withNewFeatures.csv", index=False)

        Open       High        Low      Close    Volume  Year  Month  Day  \
1  30.240000  30.240000  30.240000  30.240000   1169499  2023      6   16   
2  31.000000  31.100000  29.940001  29.940001   8064437  2023      6   20   
3  26.959999  26.959999  26.959999  26.959999   2147415  2023      6   21   
4  25.620001  27.620001  25.500000  25.940001  71898180  2023      6   22   
5  26.139999  26.600000  25.700001  25.900000  19922530  2023      6   23   

   Symbol_Encoded    Predict  5_Day_Average  10_Day_Average  \
1               0  29.940001      28.870000       28.870000   
2               0  26.959999      29.226667       29.226667   
3               0  25.940001      28.660000       28.660000   
4               0  25.900000      28.116000       28.116000   
5               0  25.900000      27.796000       27.746667   

   Daily_Price_Change  Average_Volume  Maximum_Volume  Minimum_Volume  \
1            0.099636    7.158565e+05       1169499.0        262214.0   
2           -0.009921    3.165383e+06       8064437.0        262214.0   
3           -0.099532    2.910891e+06       8064437.0        262214.0   
4           -0.037834    1.670835e+07      71898180.0        262214.0   
5           -0.001542    2.064041e+07      71898180.0       1169499.0   

   Volatility  Previous_Day_Close         RSI  
1    1.937472           27.500000  100.000000  
2    1.502842           30.240000   90.131601  
3    1.670370           29.940001   45.514943  
4    1.890048           26.959999   38.920457  
5    2.139411           25.940001   38.700562  

Standardization and stats

import pandas as pd
from sklearn.preprocessing import StandardScaler

# Veri setini yükle
df = pd.read_csv("Numeric_isedataset_withNewFeatures.csv")

# Sadece sayısal sütunları seç
numeric_columns = df.select_dtypes(include=['int64', 'float64'])

# Standartlaştırma işlemi için StandardScaler'ı kullan
scaler = StandardScaler()
standardized_data = scaler.fit_transform(numeric_columns)

# Standartlaştırılmış veriyi DataFrame'e dönüştür
standardized_df = pd.DataFrame(standardized_data, columns=numeric_columns.columns)


# Standartlaştırılmış veriyi CSV dosyasına kaydet
standardized_df.to_csv("standardized_isedataset.csv", index=False)
standardized_df

	Open	High	Low	Close	Volume	Year	Month	Day	Symbol_Encoded	Predict	5_Day_Average	10_Day_Average	Daily_Price_Change	Average_Volume	Maximum_Volume	Minimum_Volume	Volatility	Previous_Day_Close	RSI
0	-0.038546	-0.038723	-0.038380	-0.038554	-0.309809	-0.432882	-0.165266	0.043639	-1.725953	-0.037914	-0.041167	-0.044603	2.432918	-0.344201	-0.339627	-0.339898	-0.021871	-0.039540	2.347486
1	-0.038536	-0.038712	-0.038384	-0.038558	-0.121767	-0.432882	-0.165266	0.504852	-1.725953	-0.037951	-0.041161	-0.044597	-0.358623	-0.272701	-0.208493	-0.339898	-0.021924	-0.039503	1.833829
2	-0.038589	-0.038766	-0.038423	-0.038597	-0.283139	-0.432882	-0.165266	0.620155	-1.725953	-0.037964	-0.041170	-0.044607	-2.641957	-0.280129	-0.208493	-0.339898	-0.021903	-0.039507	-0.488495
3	-0.038606	-0.038757	-0.038442	-0.038610	1.619132	-0.432882	-0.165266	0.735459	-1.725953	-0.037964	-0.041178	-0.044616	-1.069858	0.122608	1.005548	-0.339898	-0.021877	-0.039548	-0.831742
4	-0.038599	-0.038770	-0.038439	-0.038611	0.201631	-0.432882	-0.165266	0.850762	-1.725953	-0.037964	-0.041182	-0.044622	-0.145135	0.237382	1.005548	-0.299326	-0.021847	-0.039562	-0.843187
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
114395	-0.038844	-0.039021	-0.038681	-0.038855	-0.254404	2.310097	-1.304996	0.043639	1.673645	-0.038201	-0.041483	-0.044967	-0.565266	-0.226983	-0.202115	-0.211732	-0.022091	-0.039817	0.400878
114396	-0.038846	-0.039022	-0.038683	-0.038857	-0.246743	2.310097	-1.304996	0.389549	1.673645	-0.038197	-0.041484	-0.044967	-0.825624	-0.224918	-0.202115	-0.208478	-0.022081	-0.039819	-0.255045
114397	-0.038847	-0.039022	-0.038683	-0.038853	-0.245244	2.310097	-1.304996	0.504852	1.673645	-0.038195	-0.041484	-0.044967	0.968180	-0.234726	-0.202115	-0.208478	-0.022083	-0.039822	0.145345
114398	-0.038845	-0.039020	-0.038681	-0.038851	-0.179529	2.310097	-1.304996	0.620155	1.673645	-0.038194	-0.041485	-0.044967	0.498292	-0.221683	-0.202115	-0.208478	-0.022085	-0.039818	0.399393
114399	-0.038843	-0.039017	-0.038678	-0.038850	-0.185710	2.310097	-1.304996	0.735459	1.673645	-0.038194	-0.041484	-0.044967	0.241299	-0.237328	-0.248774	-0.208478	-0.022079	-0.039816	0.662611

114400 rows × 19 columns

import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt

# Veri setini yükleyin
df = pd.read_csv("standardized_isedataset.csv")

# Her bir sütun için istatistiksel özet
stats_df = df.describe()

# Yeni CSV dosyasına kaydet
stats_df.to_csv("stats_isedataset.csv", index=False)

# İstatistiksel özet tablosunu göster
stats_df

	Open	High	Low	Close	Volume	Year	Month	Day	Symbol_Encoded	Predict	5_Day_Average	10_Day_Average	Daily_Price_Change	Average_Volume	Maximum_Volume	Minimum_Volume	Volatility	Previous_Day_Close	RSI
count	1.144000e+05	1.144000e+05	1.144000e+05	1.144000e+05	1.144000e+05	1.144000e+05	1.144000e+05	1.144000e+05	1.144000e+05	1.144000e+05	1.144000e+05	1.144000e+05	1.144000e+05	1.144000e+05	1.144000e+05	1.144000e+05	1.144000e+05	1.144000e+05	1.144000e+05
mean	-3.017897e-16	1.941060e-16	3.714037e-17	-2.773163e-16	2.548197e-16	1.438000e-13	-2.235353e-16	-1.019144e-16	1.420137e-14	-5.127382e-17	1.551504e-17	2.927525e-16	9.400066e-17	1.007125e-16	1.592420e-16	-1.165544e-15	-4.923582e-17	-1.490928e-17	-2.652647e-16
std	1.000004e+00	1.000004e+00	1.000004e+00	1.000004e+00	1.000004e+00	1.000004e+00	1.000004e+00	1.000004e+00	1.000004e+00	1.000004e+00	1.000004e+00	1.000004e+00	1.000004e+00	1.000004e+00	1.000004e+00	1.000004e+00	1.000004e+00	1.000004e+00	1.000004e+00
min	-3.892925e-02	-3.910575e-02	-3.876393e-02	-3.893759e-02	-3.417037e-01	-4.328823e-01	-1.589929e+00	-1.685911e+00	-1.725953e+00	-3.827785e-02	-4.157864e-02	-4.507743e-02	-1.601519e+01	-3.650962e-01	-3.618693e-01	-3.516240e-01	-2.210560e-02	-3.990638e-02	-2.857575e+00
25%	-3.881965e-02	-3.899347e-02	-3.865679e-02	-3.882808e-02	-3.279434e-01	-4.328823e-01	-1.020064e+00	-8.787876e-01	-8.744158e-01	-3.817247e-02	-4.145610e-02	-4.493711e-02	-5.932093e-01	-3.486805e-01	-3.456415e-01	-3.372884e-01	-2.207770e-02	-3.979137e-02	-7.005876e-01
50%	-3.862620e-02	-3.879402e-02	-3.846808e-02	-3.863481e-02	-2.855387e-01	-4.328823e-01	1.196667e-01	-7.166443e-02	3.322323e-03	-3.798627e-02	-4.123851e-02	-4.468878e-02	-7.462400e-02	-2.999277e-01	-2.970659e-01	-2.932099e-01	-2.202653e-02	-3.958822e-02	2.558549e-02
75%	-3.812795e-02	-3.828154e-02	-3.798377e-02	-3.813704e-02	-1.276548e-01	-4.328823e-01	9.744646e-01	8.507620e-01	8.876108e-01	-3.750627e-02	-4.068195e-02	-4.405329e-02	4.997209e-01	-1.269333e-01	-1.225789e-01	-1.309946e-01	-2.186921e-02	-3.906637e-02	7.296960e-01
max	1.157220e+02	1.156265e+02	1.157965e+02	1.157076e+02	2.634714e+01	2.310097e+00	1.544330e+00	1.773188e+00	1.673645e+00	1.110694e+02	1.145794e+02	1.090654e+02	1.387008e+01	1.893881e+01	1.824998e+01	2.278814e+01	1.355416e+02	1.218055e+02	2.347486e+00

correlation

import pandas as pd
import seaborn as sns
import numpy as np
import matplotlib.pyplot as plt

# final_dataset.csv dosyasını DataFrame'e yükleyin
df = pd.read_csv("standardized_isedataset.csv")

# Sayısal sütunları seçin
numeric_cols = df.select_dtypes(include=[np.number]).columns

# Korelasyon matrisini hesaplayın
correlation_matrix = df[numeric_cols].corr()

# Korelasyon matrisinin her bir öğesinin mutlak değerini alın
absolute_correlation_matrix = correlation_matrix.abs()

# Matristeki değerleri normalize edin (isteğe bağlı olarak)
normalized_correlation_matrix = absolute_correlation_matrix / absolute_correlation_matrix.max().max()

# Korelasyon haritasını çizin
plt.figure(figsize=(20, 10))  # Kutucukları büyütme
sns.heatmap(normalized_correlation_matrix, annot=True, cmap='coolwarm', linewidths=.5, annot_kws={"size": 12})  # Kutucuk boyutunu belirleme
plt.show()

I created a separate chart for each symbol using the new features I added and the inputs that affect the output values the most. I used Orange to see which inputs had the most impact.

import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt

# Veri setini yükleyin
df = pd.read_csv("standardized_isedataset.csv")

# 'Symbol_Encoded' sütunundaki benzersiz sembollerin listesini alın
symbols = df['Symbol_Encoded'].unique()

# Her sembol için girdileri ile 'Predict' sütununu karşılaştıran grafikler oluşturun
for symbol in symbols:
    symbol_df = df[df['Symbol_Encoded'] == symbol]
    
    # Sembol için girdileri seçin
    input_cols = ['Close', 'Low', 'High', 'RSI','5_Day_Average','10_Day_Average','Daily_Price_Change'
                  ,'Average_Volume','Maximum_Volume','Minimum_Volume','Volatility','Previous_Day_Close']
    
    # Her bir girdi sütunu için grafikler oluşturun
    for column in input_cols:
        plt.figure(figsize=(18, 5)) 
        
        # Histogram
        plt.subplot(1, 5, 1)
        sns.histplot(symbol_df[column], kde=True)
        plt.title(f"{column} Histogram")
        
        # Bar Grafik
        plt.subplot(1, 5, 2)
        sns.barplot(x='Predict', y=column, data=symbol_df)
        plt.title(f"{column} vs Prediction Bar Chart - Symbol: {symbol}")
        
        # Scatter Plot
        plt.subplot(1, 5, 3)
        sns.scatterplot(x=column, y='Predict', data=symbol_df)
        plt.title(f"{column} vs Prediction Scatter Plot - Symbol: {symbol}")
        
        # Violin Plot
        plt.subplot(1, 5, 4)
        sns.violinplot(y=column, data=symbol_df)
        plt.title(f"{column} Violin Plot - Symbol: {symbol}")
        
        # Box Plot
        plt.subplot(1, 5, 5)
        sns.boxplot(y=column, data=symbol_df)
        plt.title(f"{column} Box Plot - Symbol: {symbol}")
        
        plt.tight_layout()
        plt.show()

---------------------------------------------------------------------------
KeyboardInterrupt                         Traceback (most recent call last)
Input In [1], in <cell line: 12>()
     46 plt.title(f"{column} Box Plot - Symbol: {symbol}")
     48 plt.tight_layout()
---> 49 plt.show()

File ~/opt/anaconda3/lib/python3.9/site-packages/matplotlib/pyplot.py:368, in show(*args, **kwargs)
    324 """
    325 Display all open figures.
    326 
   (...)
    365 explicitly there.
    366 """
    367 _warn_if_gui_out_of_main_thread()
--> 368 return _backend_mod.show(*args, **kwargs)

File ~/opt/anaconda3/lib/python3.9/site-packages/matplotlib_inline/backend_inline.py:41, in show(close, block)
     39 try:
     40     for figure_manager in Gcf.get_all_fig_managers():
---> 41         display(
     42             figure_manager.canvas.figure,
     43             metadata=_fetch_figure_metadata(figure_manager.canvas.figure)
     44         )
     45 finally:
     46     show._to_draw = []

File ~/opt/anaconda3/lib/python3.9/site-packages/IPython/core/display_functions.py:298, in display(include, exclude, metadata, transient, display_id, raw, clear, *objs, **kwargs)
    296     publish_display_data(data=obj, metadata=metadata, **kwargs)
    297 else:
--> 298     format_dict, md_dict = format(obj, include=include, exclude=exclude)
    299     if not format_dict:
    300         # nothing to display (e.g. _ipython_display_ took over)
    301         continue

File ~/opt/anaconda3/lib/python3.9/site-packages/IPython/core/formatters.py:178, in DisplayFormatter.format(self, obj, include, exclude)
    176 md = None
    177 try:
--> 178     data = formatter(obj)
    179 except:
    180     # FIXME: log the exception
    181     raise

File <decorator-gen-2>:2, in __call__(self, obj)

File ~/opt/anaconda3/lib/python3.9/site-packages/IPython/core/formatters.py:222, in catch_format_error(method, self, *args, **kwargs)
    220 """show traceback on failed format call"""
    221 try:
--> 222     r = method(self, *args, **kwargs)
    223 except NotImplementedError:
    224     # don't warn on NotImplementedErrors
    225     return self._check_return(None, args[0])

File ~/opt/anaconda3/lib/python3.9/site-packages/IPython/core/formatters.py:339, in BaseFormatter.__call__(self, obj)
    337     pass
    338 else:
--> 339     return printer(obj)
    340 # Finally look for special method names
    341 method = get_real_method(obj, self.print_method)

File ~/opt/anaconda3/lib/python3.9/site-packages/IPython/core/pylabtools.py:151, in print_figure(fig, fmt, bbox_inches, base64, **kwargs)
    148     from matplotlib.backend_bases import FigureCanvasBase
    149     FigureCanvasBase(fig)
--> 151 fig.canvas.print_figure(bytes_io, **kw)
    152 data = bytes_io.getvalue()
    153 if fmt == 'svg':

File ~/opt/anaconda3/lib/python3.9/site-packages/matplotlib/backend_bases.py:2299, in FigureCanvasBase.print_figure(self, filename, dpi, facecolor, edgecolor, orientation, format, bbox_inches, pad_inches, bbox_extra_artists, backend, **kwargs)
   2297 if bbox_inches:
   2298     if bbox_inches == "tight":
-> 2299         bbox_inches = self.figure.get_tightbbox(
   2300             renderer, bbox_extra_artists=bbox_extra_artists)
   2301         if pad_inches is None:
   2302             pad_inches = rcParams['savefig.pad_inches']

File ~/opt/anaconda3/lib/python3.9/site-packages/matplotlib/figure.py:1632, in FigureBase.get_tightbbox(self, renderer, bbox_extra_artists)
   1629     artists = bbox_extra_artists
   1631 for a in artists:
-> 1632     bbox = a.get_tightbbox(renderer)
   1633     if bbox is not None and (bbox.width != 0 or bbox.height != 0):
   1634         bb.append(bbox)

File ~/opt/anaconda3/lib/python3.9/site-packages/matplotlib/axes/_base.py:4619, in _AxesBase.get_tightbbox(self, renderer, call_axes_locator, bbox_extra_artists, for_layout_only)
   4617 if self.xaxis.get_visible():
   4618     try:
-> 4619         bb_xaxis = self.xaxis.get_tightbbox(
   4620             renderer, for_layout_only=for_layout_only)
   4621     except TypeError:
   4622         # in case downstream library has redefined axis:
   4623         bb_xaxis = self.xaxis.get_tightbbox(renderer)

File ~/opt/anaconda3/lib/python3.9/site-packages/matplotlib/axis.py:1108, in Axis.get_tightbbox(self, renderer, for_layout_only)
   1105 self._update_label_position(renderer)
   1107 # go back to just this axis's tick labels
-> 1108 ticklabelBoxes, ticklabelBoxes2 = self._get_tick_bboxes(
   1109             ticks_to_draw, renderer)
   1111 self._update_offset_text_position(ticklabelBoxes, ticklabelBoxes2)
   1112 self.offsetText.set_text(self.major.formatter.get_offset())

File ~/opt/anaconda3/lib/python3.9/site-packages/matplotlib/axis.py:1085, in Axis._get_tick_bboxes(self, ticks, renderer)
   1083 def _get_tick_bboxes(self, ticks, renderer):
   1084     """Return lists of bboxes for ticks' label1's and label2's."""
-> 1085     return ([tick.label1.get_window_extent(renderer)
   1086              for tick in ticks if tick.label1.get_visible()],
   1087             [tick.label2.get_window_extent(renderer)
   1088              for tick in ticks if tick.label2.get_visible()])

File ~/opt/anaconda3/lib/python3.9/site-packages/matplotlib/axis.py:1085, in <listcomp>(.0)
   1083 def _get_tick_bboxes(self, ticks, renderer):
   1084     """Return lists of bboxes for ticks' label1's and label2's."""
-> 1085     return ([tick.label1.get_window_extent(renderer)
   1086              for tick in ticks if tick.label1.get_visible()],
   1087             [tick.label2.get_window_extent(renderer)
   1088              for tick in ticks if tick.label2.get_visible()])

File ~/opt/anaconda3/lib/python3.9/site-packages/matplotlib/text.py:910, in Text.get_window_extent(self, renderer, dpi)
    907     raise RuntimeError('Cannot get window extent w/o renderer')
    909 with cbook._setattr_cm(self.figure, dpi=dpi):
--> 910     bbox, info, descent = self._get_layout(self._renderer)
    911     x, y = self.get_unitless_position()
    912     x, y = self.get_transform().transform((x, y))

File ~/opt/anaconda3/lib/python3.9/site-packages/matplotlib/text.py:441, in Text._get_layout(self, renderer)
    438 bbox = Bbox.from_bounds(xmin, ymin, width, height)
    440 # now rotate the positions around the first (x, y) position
--> 441 xys = M.transform(offset_layout) - (offsetx, offsety)
    443 ret = bbox, list(zip(lines, zip(ws, hs), *xys.T)), descent
    444 self._cached[key] = ret

KeyboardInterrupt: 

Since I printed the graphics separately for each symbol, I printed out a few samples and stopped them.

Model

from sklearn.model_selection import train_test_split, cross_val_score
from sklearn.linear_model import LinearRegression
from sklearn.tree import DecisionTreeRegressor
from sklearn.ensemble import RandomForestRegressor
from sklearn.metrics import r2_score
import numpy as np
import joblib
import pandas as pd

# Veri setini yükle
data = pd.read_csv("standardized_isedataset.csv")

# Her bir sembolden eşit sayıda veri almak için veriyi sembollere göre grupla
grouped_data = data.groupby('Symbol_Encoded')
grouped_data = pd.DataFrame(grouped_data.apply(lambda x: x.sample(grouped_data.size().min()).reset_index(drop=True)))

# Bağımsız değişkenler ve hedef değişkeni ayır
X = grouped_data.drop(columns=["Predict"])
y = grouped_data["Predict"]

# Model tanımları
models = {
    "Simple Linear Regression": LinearRegression(),
    "Decision Tree Regression": DecisionTreeRegressor(),
    "Random Forest Regression": RandomForestRegressor()
}

best_model_name = None
best_score = -1

# Modelleri değerlendir, veri setini bölebilir ve hem ortalama R^2 hem de test R^2 değerlerini yazdır
for name, model in models.items():
    # Veri setini test etmek için bölelim
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
    
    # Çapraz doğrulama ile modeli değerlendir
    scores = cross_val_score(model, X_train, y_train, cv=10, scoring='r2')
    mean_score = np.mean(scores)
    
    if mean_score > best_score:
        best_score = mean_score
        best_model_name = name
        
    # Modeli eğitim seti üzerinde eğit
    model.fit(X_train, y_train)
    
    # Test R^2 hesapla
    y_pred = model.predict(X_test)
    r2 = r2_score(y_test, y_pred)
    
    # Eğitim R^2 hesapla
    y_pred_train = model.predict(X_train)
    r2_train = r2_score(y_train, y_pred_train)
    
    # Sonuçları yazdır
    print(f"{name}: Cross-validated R^2 scores: {scores}")
    print(f"{name}: Average Cross-validated R^2 score: {np.mean(scores)}")
    print(f"{name}: Test R^2 score = {r2}")
    print(f"{name}: Train R^2 score = {r2_train}")
    print("--------------------------------------")

# En iyi modeli seç
best_model = models[best_model_name]

# En iyi modeli eğitim seti üzerinde eğit
best_model.fit(X_train, y_train)

# En iyi modelin eğitim seti performansını hesapla
y_pred_train_best = best_model.predict(X_train)
r2_train_best = r2_score(y_train, y_pred_train_best)

# En iyi modelin test seti performansını hesapla
y_pred_best = best_model.predict(X_test)
r2_best = r2_score(y_test, y_pred_best)

# En iyi modelin adını ve performansını yazdır
print(f"The best model is: {best_model_name}, Average cross-validation R^2: {best_score}")
print(f"The best model is ({best_model_name}): Train R^2 score = {r2_train_best}, Test R^2 score = {r2_best}")

# En iyi modeli diske kaydet
joblib.dump(best_model, 'Best_Model.pkl')

Simple Linear Regression: Cross-validated R^2 scores: [0.99957118 0.99864285 0.9967277  0.99989179 0.99986738 0.99930626
 0.99892652 0.99991406 0.99987594 0.99415999]
Simple Linear Regression: Average Cross-validated R^2 score: 0.9986883660169404
Simple Linear Regression: Test R^2 score = 0.9750645913388394
Simple Linear Regression: Train R^2 score = 0.9996791190684872
--------------------------------------
Decision Tree Regression: Cross-validated R^2 scores: [0.98501935 0.99257975 0.97920722 0.99185704 0.99251042 0.9723447
 0.98772825 0.8459658  0.98899607 0.98688656]
Decision Tree Regression: Average Cross-validated R^2 score: 0.9723095155955346
Decision Tree Regression: Test R^2 score = 0.9943272051054185
Decision Tree Regression: Train R^2 score = 1.0
--------------------------------------
Random Forest Regression: Cross-validated R^2 scores: [0.9952095  0.99572558 0.99502023 0.97396329 0.99920019 0.99651583
 0.99470927 0.98587237 0.9973261  0.99526298]
Random Forest Regression: Average Cross-validated R^2 score: 0.9928805319411248
Random Forest Regression: Test R^2 score = 0.999330653305865
Random Forest Regression: Train R^2 score = 0.9984331463991017
--------------------------------------
The best model is: Simple Linear Regression, Average cross-validation R^2: 0.9986883660169404
The best model is (Simple Linear Regression): Train R^2 score = 0.9996791190684872, Test R^2 score = 0.9750645913388394

['Best_Model.pkl']